Solutions to Just KNIME It!" Challenge 12 - Season 2

:boom: New Wednesday, new Just KNIME It! challenge! :boom:

This week we have a relatively easy challenge on a topic that we haven’t explored very much: image processing! :framed_picture:

:sparkles: The goal? Preserve the identity of a crime witness by blurring their image. :male_detective:

Here is the challenge. Let’s use this thread to post our solutions to it, which should be uploaded to your public KNIME Hub spaces with tag JKISeason2-12.

:sos: Need help with tags? To add tag JKISeason2-12 to your workflow, go to the description panel on the right in KNIME Analytics Platform, click the pencil to edit it, and you will see the option for adding tags right there. :slight_smile: Let us know if you have any problems!

3 Likes

Hi all,
This is my simple solution.

3 Likes

My bit on the entirely new domain…Not sure how you define the blur… whether you maintain the color/ RGB integrity or not… Lot to learn and explore.


blur

3 Likes

I know next to nothing about image processing, so here’s an amateur’s shot:

1 Like

Hello JKIers,
Here is my take on ‘JKI S02-CH12’. For this opportunity I’ve tested a Py coding approach, based on Image SVD (singular value decomposition)

Keep KNIMEing

6 Likes

Hi all,
this is mine…similar to other solutions.
I just tried to give the possibility to choose some parameters with the slider widget and get the image from a personal URL.


7 Likes

Hello everyone,

It was an interesting exercise, since I do not have much experience working with images. Here is my solution, I found multiple way to process the image to meet the requirements and created an interactive component.

6 Likes

@Artem Great workflow

2 Likes

Hello JKIers,
As the other participants do, I’m enjoying this challenge as well; Do so, I’ve been kept working in and upgrading the solution.

So aiming to keep you updated; I’ve added a functional requirement (Blurring Method), based in Python as well. This is the classical ‘Pixelation’. A combination of the two Methods (SVD + Pixelation) would be doable, but this is enough for the challenge.

Enjoy JKI-ing

P.S. Pixelation % stands for amount of pixels.

5 Likes

I think I’m polyamorous, KNIME :orange_heart: Python have me crazy :crazy_face:

I already saw all possible solutions with de KNIME nodes hence I decide to implement all work in Python, in only one node. JKISeason2-12 Ángel Molina – KNIME Community Hub

Inspired by this workflow Scripting _ Py View _ Embedding Image from www into Table – KNIME Community Hub from my friend @gonhaddock

Looks simple, but I have had some migraine from that :sweat_smile: :rofl:

output
As always, if you are interested, here is my Linkedin and Twitter where I´m sharing other projects from KNIME.

Have a good weekend :beach_umbrella: CRAZY KNIMErs

3 Likes

Hi everyone,
Here is my solution.

1 Like

Hello everyone,
Here’s my solution.

I tried blurring the image with arbitrary height. It is possible to change the sigma and blurred area. I used “image cropper” and “merger”, which were complicated for me to use, but I managed to do it. Thanks.

8 Likes

I’m loving the fact that so many of you created data apps for users to tweak the blurring! Again you folks went the extra mile!!

2 Likes

Hi All- here is my solution. On the side, i am learning to play a bit with the data app header component we all loved last week and so i chose to use this to render the image.


Great challenge!
L

4 Likes

Hi Everyone :slight_smile:

Great workflows!!

Mine is very simple this week. I have used the -Single Selection Widget- and the -Slider Widget- nodes so the user can change the Out of Bounds Strategy and the Sigmas value in the -Gaussian Convolution- node. Afterwards, I’ve converted the ImgPlus to PNG and used the -Image Output Widget- node to display the image in the interactive view:

Here is my workflow:

Best wishes
Heather

5 Likes

:female_detective: As always on Tuesdays, here’s our solution to last week’s challenge! :female_detective:

:sparkles: We kept it very simple since this is an easy challenge. The key here is to learn how to properly configure the Gaussian Convolution node!

2 Likes

Hello, KNIME’s community,

A lot late dut the past week has a lot of work, on this challenge, I see several did attractive solutions for blurring the image.

At this time I merge two solutions in the same workflow:

  • One reads the image and applies the Gaussian Convolution component
  • The second, was implemented using R

Here is the link to the solution: JKISeason2-12 – KNIME Community Hub

image

Here are the 1st solution results:
image

and the 2nd solution results:

1 Like

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.